The Impact of AI & Data on Package Printing, Converting

Anyone who isn’t familiar with data management and AI has probably been living under a very large rock of the past few years. There is no denying their impact on every aspect of our lives, from personal interactions, to business transactions. And the label and packaging industry is no exception.

Andy Paparozzi, chief economist at PRINTING United Alliance, notes that the impact hasn’t been confined to just one area either. “Every mission-critical function from quality control to supply chain management, and from customer-preference analysis to personalization based on those preferences, benefits from artificial intelligence and superior data management — i.e., the creation, maintenance, and accessibility of relevant, robust databases.”

“I don't know if everybody realized how fast this is changing,” notes Steve Metcalf, co-founder and CEO of Imagine AI Live. “But I would start by saying that [in the printing and packaging industry] there's already a big expertise in managing data for clients. If you're printing labels for pharma, for example, you’ve got to ensure they are all labeled correctly. There's this huge amount of data and trust that goes on, that's given to printing and converting houses. It’s a perfect world for this new sort of capability coming their way with generative AI in the reasoning, and these language models we hear about and that we're all using.”

There are, Paparozzi notes, three major “buckets” that we can classify data into, and in each of those, managing that data is a given, but AI is making huge strides in exactly how that data is managed.

1. Structured data such as financials, market demographics, and customer profiles that fit neatly into conventional computer spreadsheets.
2. Unstructured data such as audio files, video files, and text files that spreadsheets do not handle well.
3. Semi-structured data such as data generated by Internet of Things (IoT) sensors, which is a mix of the two.

And AI, he stresses, can help analyze all of it, helping ensure packaging printers are making the best possible decisions across the board, instead of just relying on a “gut feeling” or guesswork.

As to where it’s going next? That’s hard to predict, especially with how rapidly the AI space is evolving, seemingly every day.

Metcalf notes that in particular, he believes the biggest change for the industry will come more from just an awareness than anything else. It is nigh on impossible to keep up with every single innovation, so there are capabilities out there many printers aren’t even aware exist. He believes that will likely be the biggest shift we’ll see in the short term — the next 12-18 months.

“When I talk to printing and packaging leaders, I think it's not to say that they're at fault. It's just they're so busy with their day jobs to be aware that you can do this now, and you don't have to necessarily wait for your MIS software provider to provide it for you. You don't have to wait for Microsoft. In fact, if you do, you're going to be at a disadvantage,” Metcalf stresses. “Whether it's a little injection of AI consulting, or AI app building, or whatever it is, you can literally start solving these problems today, like this afternoon. And I think just people aren't even aware of how fast this is happening.”

That’s not to say there’s not specific areas to be keeping an eye on. Paparozzi notes that the areas he is watching most closely for the next 12-18 month cycle revolve around innovations in “Operations — including quality control and predictive maintenance — because our markets are too competitive to pass inefficiencies to customers; automation to boost productivity and production speeds, and to overcome chronic labor shortages; market analysis and forecasting to support consistently superior decisions companywide; and cybersecurity and risk management because as one participant in PRINTING United Alliance research says so well, ‘No matter how well your business is doing or how well it is run, it can turn in an instant.’”

So what does that mean for the average label and packaging printer, and how can you prepare for something — much less implement it — if it is changing so rapidly? First both Metcalf and Paparozzi stress starting small — don’t try to do everything all at once. Pick one area of your operation and analyze it — see where the inefficiencies are. Ask the staff what tasks they hate doing, that are repetitive or frustrating. What kinds of overlaps are there with other tasks? Where is data being input multiple times to move through your systems? Those are the places where you start with AI tools — looking for ways to streamline one small piece of the puzzle.

And then, from there, you can start building on the process. What tasks or departments link to the ones you’ve just added AI analytics and automation to? Can those then be automated, or benefit from their own AI analytics tools? You can eventually have systems that hook into every part of the business, but don’t try to start off at that point, or you will quickly get overwhelmed.

Next, the old adage of “garbage in, garbage out” is even more critical when it comes to AI and data analytics. You can only get good results from having your tools go through things like inventory management, or production workflow, or predictive maintenance, or any host of tasks if the data it is being provided is good to begin with. If you’ve got incomplete data, or there are errors, or even cases where the same set of data is different in two separate systems, you’re not going to get the most out of any AI tools. If you need a place to start, before even exploring AI tools, make sure your data is clean and clear, and your policies about inputting and maintaining that data are spelled out in detail to everyone in the organization.

Finally, pick one person to be your data and AI “guru”. It could absolutely be a full-time job just keeping up with the new innovations, much less figuring out how to implement them in a logical way. Having one person be the designated go-to for questions from the staff, to make the recommendations about what the business should — and shouldn’t — try investing in, as well as to be the point person in checking to make sure what the AI spits out is accurate and correct is critical to getting it right. Depending on the size and scope of your operation, you might need to have several people tasked with different parts of this process, but the fact remains you need to be clear about who is taking responsibility for it.

On that note, Metcalf points out that the person who takes on the AI and data tasks doesn’t have to just be looking at what the company should invest in next — they can be playing with some of the tools and doing the work themselves as well.

Metcalf notes that you can “turn anybody into a data analyst now, and then spin up the apps around them to support them. And maybe it just becomes a stepping stone to getting some greater capability down the road — I use the term burner apps, like you can literally create a burner app now that gives you a taste of what you can do. And maybe that's not the end game. Maybe you spend a few hours on it and say ‘that wasn't quite right.’  But now you have a better sense of what you can do.” And that, he stresses, means that even if you do ultimately decide to start exploring a solution from different vendors, you have a much better idea of what, exactly you need, rather than just picking an AI tool at random and hoping for the best.

And the benefits of AI, especially in the data management space, are building up daily. Paparozzi notes, “Embracing AI and being data driven allows companies of all types and sizes to be more productive by automating companywide low value-added tasks they could never automate before, freeing time for activities that create the greatest value for clients, employees, and the company, and to enhance market analysis and forecasting, critical in an industry in which the gamut of opportunity is expanding but the margin for error is shrinking.”

What advice do Metcalf and Paparozzi have when it comes to diving into the AI and data management spaces?

Metcalf notes that for him, it’s all about finding a community to bounce ideas off of and get excited with. “Let's say you're in Chicago, or Kansas City, or Boston, or wherever. It doesn't matter — any size city, look up your peer groups. Local AI groups are mushrooming up everywhere. Spend time with other AI people. Figure out how other people are applying it and using it. They all have monthly meetings where AI-minded people are getting together, and they're learning from each other. They're using each other as sounding boards and for inspiration.”

For Paparozzi, it’s an emphasis on the fact that you can’t hesitate. Refusing to engage with AI is no longer an option. He says, “Take the advice of PRINTING United Alliance State of the Industry participants: ‘You can't sit on the sidelines with regard to AI because it is going to fundamentally change our business, so make sure you are learning about. It isn’t going away,’ ‘The big winners in our industry will master AI at all levels of the company,’ and ‘We feel very strongly about our organization learning about and using AI because if we don’t, our competitors are going to pass us by.’”